Serum potassium as a predictor of 28-day and in-hospital mortality in patients with sepsis: A U-shaped association

To the Editor: Sepsis, the leading cause of intensive care unit (ICU) admissions, has a mortality rate ranging from 10% to 50%.[1] Despite significant advances in understanding sepsis, it continues to be associated with acute organ failure and poses a global health challenge. Its heterogeneous nature complicates the standardization of diagnosis, treatment, and prognosis, highlighting the importance of predicting patient outcomes. Systemic inflammation frequently results in electrolyte imbalances in ICU patients. Potassium (K+), essential for cellular metabolism and membrane potential, plays a crucial role in the proper functioning of the cardiovascular, visceral, and nervous systems. Both hypokalemia and hyperkalemia can lead to severe complications. Additional research is required to elucidate the relationship between serum K+ levels and sepsis prognosis. Therefore, this study retrospectively analyzed clinical data from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database to investigate the correlation between serum potassium levels and 28-day and in-hospital mortality in sepsis, aiming to provide insights and evidence to inform future clinical practice. This study adhered to the regulations outlined in the Declaration of Helsinki. Access to the MIMIC-IV database was approved by the review committees of Beth Israel Deaconess Medical Centre and Massachusetts Institute of Technology. As a result, the Ethical Committee of the Affiliated Kunshan Hospital of Jiangsu University waived the requirement for informed consent and ethical approval. Participants were categorized into three groups based on potassium tertiles (T1–T3). Continuous variables were compared using t-tests or Kruskal–Wallis tests, while categorical variables were compared using Fisher’s exact test. Mortality rates were calculated for each tertile. Multivariate Cox regression analysis, adjusted for confounding factors, were used to construct models and evaluate the predictive value of K+ levels. Nonlinear relationships were examined with restricted cubic splines and two-piecewise Cox models. Stratified analyses were performed to evaluate consistency across subgroups. Statistical significance was set at P <0.05, using IBM SPSS 24.0 (IBM SPSS Inc., Chicago, USA) and Empowerstats 6.0 (Empower Statistics, X Medical Technology Co., Ltd, China). A cohort of 25,203 patients with sepsis was stratified based on K+ level tertiles. The average age was 65.9 ± 16.2 years, with 57.79% being male [Supplementary Table 1, https://links.lww.com/CM9/C635]. Compared to the T2 group, patients in the T1 and T3 groups exhibited higher heart rates, mean arterial pressures, and greater prevalence of liver cirrhosis, renal failure, and respiratory failure. These patients were also more likely to require continuous renal replacement therapy (CRRT), and showed elevated levels of white blood cells, blood glucose, and lactate, along with lower levels of calcium and albumin. Additionally, they experienced longer ICU stays and higher mortality rates. Patients were further categorized based on 28-day mortality [Supplementary Table 2, https://links.lww.com/CM9/C635]. In the non-survivor group, data of white blood cells, blood glucose, creatinine, lactate, sequential organ failure assessment (SOFA) score, logistic organ dysfunction score, simplified acute physiology score II, Oxford acute severity of illness score, and acute physiology score III were collected, along with longer ICU stays and higher mortality rates. Multivariable Cox regression analysis was employed to develop three models [Supplementary Table 3, https://links.lww.com/CM9/C635]. In Model II, after adjusting for various factors (age, sex, height, weight, race, marital status, hemoglobin, platelet, red blood cell, white blood cell, glucose, Na+, lactate, SOFA score, hypertension, diabetes, heart failure, renal failure, respiratory failure, cirrhosis, and malignancy), the hazard ratios (HRs) for K+ and 28-day mortality were 0.90 (95% confidence interval [CI]: 0.77–1.06, P = 0.214) at T2 and 1.15 (95% CI: 1.00–1.33, P = 0.044) at T3. For in-hospital mortality, the HRs were 0.90 (95% CI: 0.75–1.08, P = 0.277) at T2 and 1.31 (95% CI: 1.11–1.54, P = 0.001) at T3. A nonlinear relationship between K+ levels and mortality was identified. Cox proportional hazard models with restricted cubic splines and smooth curve fitting revealed a U-shaped association between K+ levels and both 28-day and in-hospital mortality [Figure 1A,B]. The inflection points for 28-day and in-hospital mortality were at 3.8 mEq/L and 3.7 mEq/L, respectively (P <0.001) [Supplementary Table 4, https://links.lww.com/CM9/C635]. After adjusting for confounders, the risks of 28-day and in-hospital mortality decreased by 20% (HR 0.80, 95% CI: 0.61–1.04, P = 0.097) and 43% (HR 0.57, 95% CI: 0.41–0.80, P = 0.001) per unit increase in K+, respectively. Conversely, K+ levels exceeding 3.8 mEq/L (HR 1.12, 95% CI: 1.04–1.21, P = 0.002) and 3.7 mEq/L (HR 1.31, 95% CI: 1.21–1.41, P <0.001) were significantly associated with increased mortality risk. The optimal K+ level range was 3.5–4.1 mEq/L for 28-day mortality and 3.5–4.2 mEq/L for in-hospital mortality (HR ≤1.00).Figure 1: Serum potassium levels and mortality risk in sepsis: threshold and stratified analyses. The association between serum potassium (K+) levels and 28-day mortality (A) and in-hospital mortality (B) in patients with sepsis, with HRs calculated using a K+ level of 3.8 mEq/L for (A) and 3.7 mEq/L for (B) as the reference. Adjustments were made for age, sex, height, weight, race, marital status, hemoglobin, platelets, red blood cells, white blood cells, glucose, sodium, lactate, sequential organ failure assessment score, hypertension, diabetes, heart failure, renal failure, respiratory failure, cirrhosis, and malignancy. Solid lines represent the estimates, while the red areas indicate the corresponding 95% CIs. Stratified analysis of serum K+ levels and their association with 28-day mortality and in-hospital mortality (C, D). CI: Confidence interval; HRs: Hazard ratios.Stratified analyses demonstrated consistent findings across subgroups based on age, sex, race, marital status, CRRT, diabetes, hypertension, heart failure, renal failure, respiratory failure, and cirrhosis [Figure 1C,D]. Notably, in sepsis patients without malignancy, the risk of 28-day mortality decreased by 13% for K+ levels ≥3.8 mEq/L compared to levels <3.8 mEq/L (P = 0.002), indicating that malignancy significantly influences the effect of K+ on mortality (P = 0.001). No significant interactions were observed between K+ levels and other stratification variables. While previous studies have established a link between dyskalemia and outcomes in sepsis, our study further characterizes a previously under-recognized U-shaped association between baseline K+ levels and both 28-day and in-hospital mortality. Importantly, our threshold effect analysis identified critical turning points at 3.8 mEq/L for 28-day mortality and 3.7 mEq/L for in-hospital mortality. These findings highlight the precise and non-linear nature of K+ as a significant predictor of mortality in sepsis, suggesting that both hypokalemia and hyperkalemia require clinical attention. K+, the primary intracellular cation, is essential for cellular functions, fluid balance, muscle contraction, and nerve impulses. Disruptions in K+ levels can lead to organ dysfunction and adversely affect prognosis. Severe hypokalemia (<2.5 mEq/L) can cause metabolic acidosis, rhabdomyolysis, renal dysfunction, muscle cramps, and arrhythmias, while hyperkalemia (>5.5 mEq/L) can result in muscle paralysis, cardiac arrest, or life-threatening arrhythmias. Abnormal serum K+ levels are associated with poor prognosis and increased mortality risk in ICU patients. Studies indicate that K+ levels exceeding 4.5 mEq/L are strong predictors of 30-day all-cause mortality, and both hypokalemia and hyperkalemia are linked to increased patient mortality, underscoring the importance of closely monitoring K+ levels.[2] Sepsis, characterized by life-threatening organ dysfunction due to a dysregulated host response to infection and systemic vasodilation-induced hypotension, is a leading cause of ICU mortality. A fundamental pathogenic mechanism in sepsis involves the activation of macrophages and cytokines due to inflammation, followed by reactive nitrogen generation, leading to shock and increased mortality risk. Understanding the pathogenesis of septic shock is crucial, with K+ receiving significant attention due to its role in electrolyte imbalances. Studies have linked K+ channels to the development of hypotension and shock in patients with sepsis, where intracellular K+ efflux induces persistent vasodilation. This hyperpolarization of cell membranes is a primary cause of decreased vascular reactivity in septic shock, explaining the limited effectiveness of vasopressors in improving microcirculatory disorders. During septic vasodilatory shock, inflammatory cytokines, elevated nitric oxide levels, disrupted calcium channels, and activated K+ channels increase vasopressor dosage requirements and heighten the risk of arrhythmias, myocardial infarction, and heart failure. In sepsis, white blood cells, particularly neutrophils, migrate to infection sites and organs such as the abdomen and lungs, increasing the risk of multiple organ dysfunction. K+ channels play a role in inducing cytokine production and inflammatory responses through transmembrane signal transduction in macrophages. Monitoring and controlling K+ levels can help mitigate inflammation, enhance cardiac and vascular function, improve organ perfusion, and reduce mortality risk in patients with sepsis. Is it sufficient to maintain serum K+ levels within the normal range for sepsis patients? Studies suggest various optimal serum K+ ranges: 3.5–4.5 mEq/L, 4.0–5.0 mEq/L, and 4.5–5.5 mEq/L.[3,4] Generally, the safe range for ICU patients is considered 3.5–5.0 mEq/L. Hessels et al[5] found it safe to maintain serum K+ between 3.5 mEq/L and 4.0 mEq/L in critically ill patients. In our study, inflection points were observed for 28-day and in-hospital mortality at 3.8 mEq/L and 3.7 mEq/L, respectively. K+ ranges of 3.5–4.1 mEq/L for 28-day mortality and 3.5–4.2 mEq/L for in-hospital mortality were associated with an HR of ≤1.0, indicating that maintaining K+ levels within 3.5–4.1 mEq/L is relatively safe. K+ below these thresholds also pose a death risk. Both high and low K+ levels increase mortality risk in patients with sepsis. Clinicians should use these insights to guide diagnosis and treatment, complementing empirical approaches. In conclusion, this study underscores the significance of K+ as a valuable predictor for assessing the risk of 28-day and in-hospital mortality among patients with sepsis, revealing a U-shaped association between serum K+ levels and mortality. Maintaining K+ levels within the range of 3.5–4.1 mEq/L was found to be safer. Future research should investigate whether interventions aimed at modulating serum K+ levels could improve clinical outcomes in these patients. Conflicts of interest None.